Hello,
Would love some help, having trouble getting gitk up and running again...
Here's the details from `git bugreport`:
What did you do before the bug happened? (Steps to reproduce your issue)
1. Open any git repository directory in a terminal.
2. Run `gitk`
What did you expect to happen? (Expected behavior)
gitk opens
What happened instead? (Actual behavior)
gitk crashes with the following error message on the command line:
objc[33662]: autorelease pool page 0x7fb1070a3000 corrupted
magic 0x00000000 0x00000000 0x00000000 0x00000000
should be 0xa1a1a1a1 0x4f545541 0x454c4552 0x21455341
pthread 0x10b020dc0
should be 0x10b020dc0
An error window also pops up saying that "wish quit unexpectedly."
What's different between what you expected and what actually happened?
It shouldn't crash.
Anything else you want to add:
I'm using git/git-gui as installed by brew on OSX>
If I run `gitk` where outside of a git repository, then it opens fine
(and shows the "Cannot find a git repository here" error.
gitk has run fine on this computer for years; I don't know what
suddenly changed to make it stop working.
[System Info]
git version 2.30.0
cpu: x86_64
no commit associated with this build
sizeof-long: 8
sizeof-size_t: 8
shell-path: /bin/sh
uname: Darwin 19.6.0 Darwin Kernel Version 19.6.0: Tue Nov 10 00:10:30
PST 2020; root:xnu-6153.141.10~1/RELEASE_X86_64 x86_64
compiler info: clang: 12.0.0 (clang-1200.0.32.28)
libc info: no libc information available
$SHELL (typically, interactive shell): /bin/zsh
[Enabled Hooks]
I have managed to figure out how to fix the issue - I removed my
~/.config/git/gitk. Upon recreation, everything worked fine.
I ran a diff between the old and the new file and discovered the issue
was this line in the gitk configuration:
set geometry(state) zoomed
If you change "zoomed" to "normal" then gitk works again.
I don't know enough about git-gui internals to help more than this,
but it looks like the "zoomed" state is broken somehow.
-Dan
On Mon, Jan 25, 2021 at 1:57 PM Dan Lew [off-list ref] wrote:
Hello,
Would love some help, having trouble getting gitk up and running again...
Here's the details from `git bugreport`:
What did you do before the bug happened? (Steps to reproduce your issue)
1. Open any git repository directory in a terminal.
2. Run `gitk`
What did you expect to happen? (Expected behavior)
gitk opens
What happened instead? (Actual behavior)
gitk crashes with the following error message on the command line:
objc[33662]: autorelease pool page 0x7fb1070a3000 corrupted
magic 0x00000000 0x00000000 0x00000000 0x00000000
should be 0xa1a1a1a1 0x4f545541 0x454c4552 0x21455341
pthread 0x10b020dc0
should be 0x10b020dc0
An error window also pops up saying that "wish quit unexpectedly."
What's different between what you expected and what actually happened?
It shouldn't crash.
Anything else you want to add:
I'm using git/git-gui as installed by brew on OSX>
If I run `gitk` where outside of a git repository, then it opens fine
(and shows the "Cannot find a git repository here" error.
gitk has run fine on this computer for years; I don't know what
suddenly changed to make it stop working.
[System Info]
git version 2.30.0
cpu: x86_64
no commit associated with this build
sizeof-long: 8
sizeof-size_t: 8
shell-path: /bin/sh
uname: Darwin 19.6.0 Darwin Kernel Version 19.6.0: Tue Nov 10 00:10:30
PST 2020; root:xnu-6153.141.10~1/RELEASE_X86_64 x86_64
compiler info: clang: 12.0.0 (clang-1200.0.32.28)
libc info: no libc information available
$SHELL (typically, interactive shell): /bin/zsh
[Enabled Hooks]
On Tue, Jan 26, 2021 at 2:50 PM Dan Lew [off-list ref] wrote:
I have managed to figure out how to fix the issue - I removed my
~/.config/git/gitk. Upon recreation, everything worked fine.
Not that I know much about the internals of the objc runtime,
but this indicates a bug in wish and/or the objc runtime
(tickled *by* the geometry state line, but not in gitk itself).
Note that the corrupted autorelease pool page data:
magic 0x00000000 0x00000000 0x00000000 0x00000000
should be 0xa1a1a1a1 0x4f545541 0x454c4552 0x21455341
suggests that the page was already released or something,
since the three words of "should be" spell out "otua eler !esa"
which is "autorelease!" with byte-swapping (the old NUXI
issue but expressed in 32 bits this time).
Chris